I need detect object in series of pictures. This object has something like nonsymetric relief structure with some irregularities. In this conditions.

  1. Light conditions can change from almost dark to overlight (sometimes from only one side).
  2. Object can be in different sizes, i. e. I am interested in scale invariant shape of that object.
  3. Object can be sometimes badly focused. i. e. contours are sometimes hard to find and identify.
  4. Object can be turned along vertical axis, but obejct is symmetric along this axis.
  5. Sometimes part of this object can be hidden behind some other object.

Is there some way how to construct general, robust algorithm for this?

Thanks in advance

  • $\begingroup$ There is a whole field dedicated to this. Have you looked at the literature yet? $\endgroup$ – geometrikal Aug 23 '15 at 11:57
  • $\begingroup$ You could start with some of the sift based methods $\endgroup$ – geometrikal Aug 23 '15 at 11:57

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.